latest ROMS version problem

[wpgong]

I downloaded the latest version of ROMS (741) and compiled in our cluster, when i ran the model, the model stopped soon.
the problem is shown as the follows:

oceanM: string.c:177: NC_check_name: Assertion `name != ((void *)0)' failed.
forrtl: error (76): Abort trap signal
Image PC Routine Line Source
oceanM 0000000000A5E5B9 Unknown Unknown Unknown
oceanM 0000000000A5CF30 Unknown Unknown Unknown
oceanM 0000000000A0E1C2 Unknown Unknown Unknown
oceanM 00000000009CA403 Unknown Unknown Unknown
oceanM 00000000009CE7A2 Unknown Unknown Unknown
libpthread.so.0 00002B4BC84DB4A0 Unknown Unknown Unknown
libc.so.6 0000003985832885 Unknown Unknown Unknown
libc.so.6 0000003985834065 Unknown Unknown Unknown
libc.so.6 000000398582B9FE Unknown Unknown Unknown
libc.so.6 000000398582BAC0 Unknown Unknown Unknown
oceanM 00000000009A3749 Unknown Unknown Unknown
oceanM 00000000009A8337 Unknown Unknown Unknown
oceanM 00000000009B149C Unknown Unknown Unknown
oceanM 000000000089531F Unknown Unknown Unknown
oceanM 0000000000758AF1 Unknown Unknown Unknown
oceanM 000000000074BF6A Unknown Unknown Unknown
oceanM 00000000004E9833 Unknown Unknown Unknown
oceanM 00000000004B9DB6 Unknown Unknown Unknown
oceanM 0000000000418214 Unknown Unknown Unknown
oceanM 000000000041801D Unknown Unknown Unknown
oceanM 0000000000417DA6 Unknown Unknown Unknown
libc.so.6 000000398581ECDD Unknown Unknown Unknown
oceanM 0000000000417C99 Unknown Unknown Unknown
yhrun: error: cn448: task 0: Aborted
yhrun: First task exited 60s ago
yhrun: tasks 1-191: running
yhrun: task 0: exited abnormally
yhrun: Terminating job step 76154.0

could anybody help to tell me what's the problem and how to resolve?
thanks.

[kate]

NetCDF is checking for a null string somewhere and it found one. It could be a file name or it could be a variable name - hard to tell from what you've shown us.

[wpgong]

Hi, Kate:
i uploaded the ocean_estuary.in and the output.dat file from the model run.
i have the following input files:
*grd.nc, *ini.nc, *clm.nc, *oa.nc, *frc.nc, *bry.nc, stations.in, varinfo.dat
do i still need other files or variables for the model run? for the previous version of ROMS, these files are enough.
thanks.

Wenping 10/18

[jfreeman]

Hi,

I have the same issue with the model exiting due to the netcdf assertion failing. After a little bit of investigation the issue seems to be (for me at least) with writing a restart file AND having WET_DRY activated. Model runs without WET_DRY run to completion, whilst runs with WET_DRY crash with the netcdf error.

The error is this:

Code: Select all

oceanM: dstring.c:176: NC_check_name: Assertion `name != ((void *)0)' failed.
forrtl: error (76): Abort trap signal

Compiling with debug flags gave me this error:

Code: Select all

forrtl: severe (408): fort: (3): Subscript #1 of the array VID has value 0 which is less than the lower bound of 1

Image              PC                Routine            Line        Source             
oceanG             000000000138EEAA  Unknown               Unknown  Unknown
oceanG             000000000138D9A6  Unknown               Unknown  Unknown
oceanG             000000000134BBF0  Unknown               Unknown  Unknown
oceanG             000000000130738F  Unknown               Unknown  Unknown
oceanG             00000000013078B1  Unknown               Unknown  Unknown
oceanG             0000000000E08601  def_rst_                  214  def_rst.f90
oceanG             00000000006EFB4C  output_                   244  output.f90
oceanG             00000000006DBF37  main2d_                   175  main2d.f90
oceanG             000000000041C3C9  ocean_control_mod         173  ocean_control.f90
oceanG             000000000041AD28  MAIN__                    108  master.f90
oceanG             000000000041A62C  Unknown               Unknown  Unknown
libc.so.6          00002AD4DCA77D5D  Unknown               Unknown  Unknown
oceanG             000000000041A529  Unknown               Unknown  Unknown

and the code from def_rst.f90 is:

Code: Select all

status=def_var(ng, iNLM, RST(ng)%ncid, RST(ng)%Vid(idPwet),     &
 &                 NF_FOUT, nvd3, sp2dgrd, Aval, Vinfo, ncname,     &
 &                 SetFillVal = .FALSE.)
 IF (exit_flag.ne.NoError) RETURN

There might be a problem with the value of idPwet but I have not looked any further into this right now. The model runs if I make the restart interval larger than the run length.

Regards,

j

Here is the svn info

Code: Select all

$ svn info
Path: .
URL: https://www.myroms.org/svn/src/trunk
Repository Root: https://www.myroms.org/svn/src
Repository UUID: f091316a-d328-0410-a40a-876eff57d070
Revision: 741
Node Kind: directory
Schedule: normal
Last Changed Author: arango
Last Changed Rev: 741
Last Changed Date: 2014-10-16 08:05:43 +1100 (Thu, 16 Oct 2014)

[wpgong]

Hi, Freeman:
i tried your way, it works. many thanks.

Wenping

[arango]

You are getting this error because you are using an older version of the metadata file varinfo.dat. You need to use the updated file from the latest version of the code. Notice that I added to varinfo.dat the metadata for the wetting and drying mask at PSi-points that it is needed for restart:

Code: Select all

'wetdry_mask_psi'                                  ! Input/Output
  'wet/dry mask on PSI-points'
  'nondimensional'
  'wetdry_mask_psi, scalar, series'
  'ocean_time'
  'idPwet'
  'p2dvar'
  1.0d0

This will force mod_ncparam.F to initialize correctly index iPwet.

[jfreeman]

updating the varinfo.dat file fixed it. Thanks for the info!

Regards,

j